neurips18_hierchical_image_manipulation
neurips18_hierchical_image_manipulation copied to clipboard
xcyan
Pytorch Implementation of NeurIPS'18 paper on Generative Image Manipulation with Hierarchical Semantic Structures
HELLO DEAR @honglaklee @thomasehuang @xcyan @maga33 I THINK THE WAY YOU EXPECTED THE ADE20K DATASET IS NOT HANDFUL ANYMORE. I downloaded the big data set and ran your preprocess code...
hello dear in Inference using a Pre-trained Box-to-Layout Generator section, when we run the bash codes, sites like : https://umich.box.com/public/static/mi64jtkw4n5x2wgyjsznt9cm6vjgsmbp.pth don't open.
The provided ADE20K dataset link is invalid now, and the ADE20K dataset in http://sceneparsing.csail.mit.edu/ seems to be not compatible with the pre-processing code. Would mind share the processed dataset? Thanks....
How could i get the pretrained model from your git?! (uploaded URL links are forbidden...)
I dowloaded pretrained model for city dataset and when I try to run Joint Inference, it cannot find 'network' in self.netG.params_dict. I have the following error; Traceback (most recent call...
Could you please provide more information about the following options: --use_bbox --load_image --label_nc and --output_nc (should it be total number of classes or 0 for only using semantic segmentation information?)...
When I try to train cityscape dataset with --no_instance and --label_nc 0 tobe able to use only semantic segmentation information, I am getting following error. Could you please help me...